For the e-published version of this thesis,

© Kristina Berglund ISSN: 1101-718X

ISBN: 978-91-628-7669-2

ISRN: GU/PSYK/AVH--213—SE

For the e-published version of this thesis,

please visit: http://hdl.handle.net/2077/18796

Printer: Intellecta AB V Frölunda

ABSTRACT

Berglund, Kristina (2009). Socially stable alcoholics: What characterises them? Drinking patterns, personality and health aspects of psychosocial and clinical importance. Department of Psychology, University of Gothenburg, Sweden

People who misuse alcohol are a heterogeneous group with different etiology, social and clinical characteristics. This thesis includes four studies whose aim was to characterise so called socially stable alcoholics (i.e. individuals with preserved psychosocial functioning) regarding personality, physical and mental health and drinking patterns. Specifically the aim of Study I was to describe demographic and clinical characteristics in male individuals with excessive alcohol intake (n = 367) recruited by advertisements and to compare these individuals according to their prior experience of treatment. The results showed that individuals with no prior treatment history (n = 238) were found to be more often cohabitant and employed. They also reported fewer on-going psychiatric symptoms than individuals with prior treatment histories. The aim of

central serotonergic neurotransmission and years of excessive alcohol intake in 33 alcohol- dependent male individuals. More individuals with low serotonergic neurotransmission as well as long time-period of excessive drinking had elevated levels of anxiety proneness. Long time-period of excessive drinking was the strongest predictor for anxiety proneness. The aim of

in comparison to a population-based male control group (n = 131). There were differences between the two groups although the differences were mainly small to moderate in magnitude. Further analyses showed a more heterogeneous pattern among alcoholics than controls in the personality traits impulsiveness and anxiety proneness. The aim of

was to investigate alcohol-dependent men and women (n = 125) in an inpatient treatment setting and analyse specific characteristics such as substance use and health functioning in these individuals. Gender and two age-groups were compared, respectively, in these characteristics. Differences between genders were found in drinking patterns whereas differences between the two age-groups (29-47 years and 49-69 years) were found in drinking patterns as well as in somatic and mental health. The four studies reveal that there are several important characteristics among socially stable alcoholics that are related to personality, mental and somatic health, drinking patterns as well as treatment experience, barriers for help- seeking and serotonergic neurotransmission. Furthermore, the results suggest that one way to reach these alcoholics could be via alcohol treatment programs at working places and centres of learning and not via the health care system since many of them are less likely to seek treatment for psychiatric and somatic symptoms. It may also be of importance to have an age- perspective in treatment planning for alcohol-dependent individuals, where younger individuals need more of psychiatric consultations whereas their older counterparts instead need more of consultations by medical professionals.

Key words: age, alcoholics, clinical characteristics, social stability, demography, gender, health functioning, personality, serotonin, substance use

Kristina Berglund, Department of Psychology, University of Gothenburg, Box 500, S-405 30 Gothenburg, Sweden. Phone: +46 31 786 18 78, Fax: +46 31 786 46 28

E-mail: Kristina.Berglund@psy.gu.se

PREFACE

This thesis is based on the following four papers, which are referred to by roman numerals:

I. Berglund, K., Fahlke, C., Berggren, M., Eriksson, M., Balldin, J.

(2006). Individuals with excessive alcohol intake recruited by advertisement: demographic and clinical characteristics. Alcohol &

Alcoholism, 41, 200-204.

II. Berglund, K., Fahlke, C., Berggren, U., Eriksson, M., Balldin, J.

(2006). Personality profile in type I alcoholism: long duration of alcohol intake and low serotonergic activity are predictive factors of anxiety proneness. Journal of Neural Transmission, 113, 1287-1298.

III. Berglund, K., Roman, E., Balldin, J., Berggren, U., Eriksson, M., Gustavsson, P., Fahlke, C. Personality profile in socially stable men with excessive alcohol intake. Manuscript.

IV. Berglund, K., Balldin, J., Berggren, U., Fahlke, C. (2008). Self-

reported substance use and health functioning in Swedish alcohol-

dependent individuals: age and gender perspectives. Nordic Journal

of Psychiatry, 62, 405-412.

POPULÄRVETENSKAPLIG SVENSK SAMMANFATTNING

En ansenlig del av befolkningen, både i Sverige och runt om i världen har en så hög alkoholkonsumtion att det på sikt medför problem både för den enskilda individen och för samhället. De skador som alkoholproblem kan orsaka är av både kronisk art (t.ex. leverskador) och akut art (t.ex. trafikolyckor). De samhällsekonomiska kostnaderna är omfattande då alkoholproblem ger ökade kostnader för bl.a. sjukvård, socialtjänst och rättsväsende.

Personer med hög och därmed problematisk alkoholkonsumtion är en stor men samtidigt heterogen grupp i samhället. I en nyligen genomförd undersökning av Folkhälsoinstitutet (2008) fann man att cirka 17 procent av männen och ungefär 10 procent av kvinnorna tillhör denna grupp. En del utvecklar ett fysiologiskt och/eller psykologiskt beroende. Konsekvenserna för dessa, exempelvis inom arbetslivet, blir många gånger betydande. De faktorer som medverkar till att alkoholproblem utvecklas kan vara många. Till exempel kan vissa

personlighetsdrag, så som impulsivitet eller ängslighet öka sårbarheten. Även psykisk ohälsa, exempelvis depressiva symtom och ångest, kan öka sårbarheten för att utveckla alkoholproblem.

I den här doktorsavhandlingen har så kallade socialt stabila personer med

samtidig problematisk alkoholkonsumtion studerats. Med begreppet social

stabilitet menas individer som har en social förankring i samhället, så som eget

boende, samt att de är en del av den arbetsföra befolkningen. Avhandlingen är

uppdelad i fyra vetenskapliga rapporter där varje studie har som målsättning att

karakterisera socialt stabila individer med alkoholproblem utifrån några

psykologiska och psykosociala aspekter som exempelvis personlighet, psykisk hälsa och dryckesvanor.

I studie I intervjuades 367 män. Dessa hade besvarat en annons som var avsedd för personer som ansåg sig ha alkoholproblem och som ville vara med i en studie där medicinsk behandling skulle prövas. Alla deltagarna kunde beskrivas som socialt stabila och majoriteten hade aldrig tidigare fått behandling för sina alkoholproblem (70 %). De övriga hade haft någon form av kontakt med öppenvården (t.ex. primärvård; 20 %) eller med slutenvården (t.ex. psykiatrisk klinik eller behandlingshem; 10 %).

Resultaten visade att de individer som inte tidigare fått behandling hade en lika allvarlig problematisk alkoholkonsumtion som dem med erfarenhet av

öppenvården samt slutenvården. Av dem som inte hade någon tidigare behandlingserfarenhet fanns det fler som mådde psykiskt bättre, levde i en relation och hade ett arbete jämfört med dem som hade erfarenhet av

slutenvården. Vidare visade resultaten att förvärvsarbete och låg grad av psykisk ohälsa minskade sannolikheten för att söka professionell vård.

I studie II undersöktes om det fanns en relation mellan olika

personlighetsegenskaper, så som impulsivitet och ängslighet, och antal år med problematisk alkoholkonsumtion samt om det fanns en relation till hjärnans transmittorsubstans serotonin. Denna substans anses ha betydelse för bl.a.

stämningsläge, impulskontroll och ängsligt beteende. De som ingick i undersökningen var 33 socialt stabila män med ett diagnostiserat alkoholberoende.

I studien framkom det att nedsatt serotoninaktivitet var förenat med ängslighet

och att individer med långvarig och problematisk alkoholkonsumtion (mer än 9

år) hade en ängsligare läggning jämfört med dem som hade kortare tid av alkoholproblem. Personer med ett långvarit drickande hade också mer av utanförskapskänslor. Eftersom fler individer med nedsatt serotoninfunktion och/eller långvarig alkoholkonsumtion uppvisade ängsliga personlighetsdrag gjordes ytterligare en analys för att undersöka vilken av dessa variabler som var starkast förenade med detta personlighetsdrag. Analysen visade att det som mest påverkade personlighetsdraget ängslighet var långvarigt problematiskt

drickande.

Syftet med den tredje studien (studie III) var att ytterligare fördjupa kunskapen om socialt stabila individers personlighetsegenskaper. I studien ingick 100 män med alkoholproblem och en kontrollgrupp på 131 män. Resultaten visade att majoriteten av individerna i båda grupperna låg inom det område som anses vara normalvariationen för ett antal olika egenskaper. I studien användes den

statistiska metoden ’principalkomponentanalys’ som utifrån olika faktorer undersöker om det finns en systematisk struktur i data (t.ex. i

personlighetsegenskaper). Analysen visade att det fanns en större variation i hur de olika personlighetsegenskaperna kom till uttryck bland individer med

alkoholproblem. Det fanns procentuellt sett fler personer med alkoholproblem vars personlighetsegenskaper låg utanför den så kallade normalvariationen än jämfört med kontroller. Framförallt fanns det fler personer med ängsligt

personlighetsdrag, impulsiv läggning, men också frånvaro av impulsivitet, bland dem med alkoholproblem.

I den sista studien (studie IV) intervjuades 125 personer med ett

alkoholberoende, såväl kvinnor som män, vilka deltog i en behandling på ett

behandlingshem. Majoriteten av individerna var socialt stabila i den mening att

de hade arbete och bostad. Många hade dock erfarenhet av både fysisk och

psykisk ohälsa. Individerna jämfördes dels utifrån kön och dels utifrån ålder (29-

47 år jämfört med 49-69 år) på ett antal faktorer så som psykosocial bakgrund (utbildning, relationer, arbetsförhållanden, bostadsförhållanden), psykisk och fysisk hälsa samt dryckesmönster.

Vid jämförelse mellan könen fanns inga skillnader vad gäller den psykosociala bakgrunden eller i deras psykiska och fysiska hälsa. Däremot framkom vissa skillnader i dryckesmönstret; män hade t.ex. i genomsnitt provat alkohol i tidigare ålder och de hade fått problem med alkohol i yngre åldrar. De hade också haft ett genomsnittligt problematiskt drickande under längre tid än

kvinnorna. När de två åldersgrupperna jämfördes återfanns inga skillnader i den psykosociala bakgrunden, men däremot framkom skillnader i deras psykiska och fysiska hälsa. Den yngre åldersgruppen hade betydligt högre grad av psykisk ohälsa medan den äldre gruppen hade mer kroppslig ohälsa. Den yngre gruppen hade också börjat dricka tidigt och även utvecklat ett problematiskt drickande tidigt (i genomsnitt vid 23 års ålder) än den äldre åldersgruppen (i genomsnitt vid 35 års ålder).

Sammanfattningsvis visar avhandlingens fyra studier att det finns en grupp människor med allvarliga alkoholproblem som dröjer länge med att söka professionell vård för sina problem. De faktorer som framförallt tycks fördröja varför individer söker vård är att man har en stabil psykosocial livssituation och att man upplever sig vara psykiskt frisk. I avhandlingen framkom också att ju längre den problematiska alkoholkonsumtionen pågått desto mer problem utvecklas, som exempelvis fysisk och psykisk ohälsa och olika svårigheter med arbete och relationer. Även personligheten påverkas av långvarig

alkoholkonsumtion; ju fler år av problematiskt drickande desto mer uttalad

ängslighet och känslor av utanförskap som följd. Vidare visar resultaten att

graden av fysisk och psykisk ohälsa var lika mellan vårdsökande socialt stabila

kvinnor och män med tung alkoholproblematik. Däremot lider yngre personer i

högre grad av psykisk ohälsa, som t.ex. depressiva symtom och ångest men också av minnes- och koncentrationssvårigheter.

Det är av stor vikt att aktivt arbeta med tidiga interventioner och åtgärder för att förebygga alkoholproblem och därmed främja folkhälsan. Det är särskilt

angeläget att nå ungdomar och unga vuxna då dessa kommer att bli en del av

den arbetsföra befolkningen. Vidare är det angeläget att på ett tidigt stadium

upptäcka dem som har utvecklat ett alkoholproblem, men som ännu inte har sökt

vård – dvs. har ett ”dolt” alkoholproblem. En möjlig väg att nå socialt stabila

personer med samtidigt alkoholproblem är att via arbetsplatser, men också på

olika lärosäten, uppmärksamma och kontinuerligt föra en dialog om alkohol-

och narkotikafrågor.

TACK!

Jag vill rikta ett stort tack till mina båda handledare professor Claudia Fahlke och docent Ulf Berggren, samt till docent Jan Balldin! Ni har alla väglett mig på det bästa av alla tänkbara sätt under min vetenskapliga resa. Ni har alltid stöttat mig, hjälpt mig och framförallt har ni gjort mig trygg! Med tryggheten som jag alltid har känt har jag med lugn och tillit fått möjlighet att prova mina

vetenskapliga vingar och låta kreativiteten få blomma. Ni har inte bara varit fantastiska mentorer under min doktorandtid, ni är också alla oerhört varma och generösa människor! Jag är glad att få ha er som vänner!

Jag vill också tacka mina andra medförfattare Matts Eriksson, Erika Roman och Petter Gustavsson som har gett mig viktiga och konstruktiva synpunkter på artiklarna.

För värdefull hjälp i den stora statistikdjungeln som jag ständigt försöker förstå mig på så vill jag rikta ett extra tack till min fina bror Tomas Berglund. Tack för att du alltid tar dig tid att lyssna och att du alltid har så bra svar!

Jag vill tacka Ann-Sophie Lindqvist för all vänskap under årens lopp och för den fantastiska omtanke jag mötte när jag var nybörjardoktorand! Jag vill också tacka Karin Allard för härlig vänskap med trevliga lunchstunder och många samtal! Tack till Elisabeth Punzi för att du ger mig så mycket inspiration och idéer, både vetenskapliga och på helt andra plan i livet! Tack till Mattias Gunnarsson och Solveig Olausson för bra vetenskapliga samtal! Jag vill också tacka Kathleen Wadell, Anne-Ingeborg Berg, Carolina Lunde, Kristina

Holmquist, Anneli Goulding, Inga Tidefors, Karin Strid, Maria Larsson, Cecilia Jakobsson, Jakob Åsberg, Kerstin Watson Falkman och Magnus Roos för trevliga samtal och glada skratt! Jag vill tacka Erik Aronsson för hjälp med intervjuer, Nadja Fahlke för databearbetning och Jörgen Engel för att ha bidragit med medel till forskningen. Ett tack till Morgan Karlsson och personal vid Aleforsstiftelsen samt Christer Andersson och personal vid Respons

Alkoholrådgivning för gott samarbete. Tack även till alla andra underbara vänner som jag har!

Tack även till Psykologiska institutionen och alla medarbetare som arbetar här!

Detta är en underbar arbetsplats att få jobba på och jag är väldigt tacksam att jag har fått den möjligheten.

Tack till alla mina nära och kära! Tack mor och far för att ni alltid har trott på

mig och uppmuntrat mig! Tack Börje och Lillemor för att ni alltid finns där och

hjälper till i alla tänkbara och otänkbara sammanhang! Tack mina kära systrar

Eva och Anna-Karin och era familjer för att ni finns. Tack Karin, Annelie och Raiath för all vänskap och generositet! Tack till alla andra släktingar både på min egen och på Björns sida!

Tack min älskade Björn som står ut med mig och som alltid ställer upp för mig och tänker på mig! Ditt hjärta är av guld! Tack också mina fina barn Adam och Axel som ger mig energi och håller mitt barnasinne levande!

Slutligen, stort tack till alla de människor med alkoholproblem som jag har fått möjlighet att träffa under årens lopp. Utan er hjälp och er frikostighet att dela med er av upplevelser och erfarenheter från era liv hade jag inte kunnat göra det som jag har gjort!

Forskningen har finansierats med hjälp av medel från Eva och Oscar Ahréns stiftelse, Sigurd och Elsa Goljes minne, Helge Ax:son Johnsons stiftelse, Adlerbertska forskningsstiftelsen, Kungliga Vetenskaps- och Vitterhets- samhället i Göteborg, Riksbankens Jubileumsfond, Arbetsmarknadens försäkringsaktiebolag (AFA), Systembolagets råd för alkoholforskning och FoU-bidrag Västra Götalandsregionen.

Göteborg, December, 2008

CONTENTS

Introduction... 14

Gender ...16

Drinking patterns...18

Psychiatric co-morbidity ...20

Antisocial Personality Disorder Depression Anxiety Disorders Personality ...24

Neuroticism/Harm Avoidance Psychoticism/Novelty Seeking Personality, alcoholism and serotonin Heredity ...30

Type 1 and type 2 alcoholism Statistically developed categorisations ...33

Type A and type B alcoholism Other categorisations Socially stable alcoholics – what characterises them? ...36

Aim ... 38

Methods ... 39

Participants ...39

Controls ...40

Procedures and instruments...40 Study I

Study II

Study III

Study IV

Statistics...43

Results... 45

Study I...45

Study II ...48

Study III...51

Study IV...54

Discussion... 59

Study I...59

Study II ...60

Study III...62

Study IV...66

Conclusion ... 70

References... 74

Appendix... 96

Introduction

Alcohol misuse is a widespread phenomenon in many countries and it is well- known that excessive alcohol consumption is related to increased illness and death. More than 60 different medical conditions and approximately four

percent of the global burden of disease are related to alcohol. Medical conditions caused by alcohol drinking are thus globally one of the most common causes of disability and death (Room, Babor & Rehm, 2005).

The costs of alcohol are large for the society. Compared with tobacco or illicit drugs, the costs of alcohol consumption are even higher due to the adverse consequences that are generated. Excessive alcohol consumption has both chronic (e.g. liver cirrhosis) and acute (e.g. car accidents) effects on health, which cause costs for the health care, welfare and judicial systems. Besides, there are also costs for the workplaces due to, for example, reduced efficacy in work-tasks, more accidents in work and premature death (Klingemann & Gmel, 2001) and more sickness absence (Hensing & Wahlström, 2004). For the abuser, but also for the relatives, excessive alcohol consumption can cause many

negative effects in everyday life, such as somatic and psychiatric symptoms, problems in family and relationships, and economical and legal problems (Gmel

& Rehm, 2003).

The condition of alcohol misuse can vary in many ways and the divergent terminologies for characterising adverse drinking pattern is a manifestation of this. Different terminologies are for example binge drinking, heavy or excessive drinking, alcohol misuse, alcohol abuse, alcoholism and alcohol-dependence.

The latter terminology alcohol-dependence is an international diagnosis which is described both in the International Classification of Diseases (ICD, 10

revision; World Health Organization, 1992) and in the fourth edition of the

Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994). Table 1 presents the criteria for substance dependence according to DSM-IV.

Table 1

The DSM-IV criteria for substance dependence 1) Tolerance, as defined by either or the following:

a. A need for markedly increased amounts of the substance to achieve intoxication or desired effect

b. Markedly diminished effect with continued use of the same amount of the substance

2) Withdrawal, as manifested by either or the following

a. The characteristic withdrawal syndrome for the substance

b. The same (or closely related) substance is taken to relieve or avoid withdrawal symptoms

3) The substance is often taken in larger amounts or over a longer period than was intended

4) There is a persistent desire or unsuccessful efforts to cut down or control substance use

5) A great deal of time is spent in activities necessary to obtain the substance, or recover from its effects

6) Important social, occupational, or recreational activities are given up or reduced because of substance use

7) The substance is continued despite knowledge of having a persistent or recurrent physical or psychological problem that is likely to have been caused or

exacerbated by the substance

Note. Three (or more) of the criteria shall occur at any time in the same 12-months period.

Alcohol dependence is a serious disorder characterised by physical and/or

psychological dependence of alcohol. It is considered as a unitary disorder, but

the etiology of alcohol dependence is multifactorial. Moreover, the risk-factors

that are contributing to the development of alcohol-dependence are different in

various groups of individuals. In addition, alcohol-dependent individuals also differ in their characteristics, as for example in drinking patterns, severity of abuse and in the degree of psychiatric co-morbidity (Babor, 1994).

Since the beginning of Jellineks classification of five types of alcoholics (Jellinek, 1960), there have been several attempts to distinguish distinct categories of alcohol-dependent individuals. Mainly, the categorisations have been made on the basis of etiology and/or different characteristics. The development of new statistical methods, especially multivariate analysis, has helped many researchers to also incorporate more complex categorisations that include clusters of defined variables. A good example of a statistical derived categorisation is the type A and type B alcoholics (Babor et al., 1992). In the following there will be a short presentation of the most common categorisations of alcohol-dependent individuals that have been made up to this date: gender, drinking patterns, psychiatric co-morbidity, personality, heredity and statistically developed categorisations.

In this thesis the term alcoholism is used for describing all forms of problematic use of alcohol. For describing individuals with problematic use of alcohol the term alcoholics is used, if not otherwise stated.

Gender

Throughout the history of alcohol research, alcohol-dependent men and women

have constantly been regarded as two differentiated groups with different

etiology and different clinical characteristics. However, specific research

relating to etiology of alcoholism, drinking patterns and treatment needs in

women has, compared to men, lagged behind (Del Boca, 1994). Nevertheless, it

is well known that there exist some differences between men and women in drinking behaviour. For example, women with excessive drinking, drink in average less alcohol than men and their age at onset of first drinking and first intoxication occurs normally later (Miller & Cervantes, 1997). However, some studies have shown that the differences between male and female drinking behaviour have decreased during the last decade (McPherson, Casswell &

Pledger, 2004; Nelson, Heath, & Kessler, 1998).

It is well established that excessive alcohol intake has different biological effects in men and women, with women being more vulnerable for alcohol-related physical illnesses (Ely, Hardy, Longford & Wadsworth, 1999; Ma, Baraona, Goozner & Lieber, 1999; Saunders, Davis & Williams, 1981) and even brain atrophy (Mann et al., 2005). Several studies also indicate that there is a faster progression of the developmental events leading to alcohol-dependence in women (Piazza, Vrbka & Yeager, 1989; Randall et al., 1999). This

phenomenon, the so called telescoping effect, is described as a more rapid progression towards psychiatric, medical and social consequences in alcohol- dependent women compared to alcohol-dependent men (Piazza et al., 1989;

Randall et al., 1999; Schuckit, Anthenelli, Bucholz, Hesselbrock & Tipp, 1995).

Due to this faster progression of adverse consequences, women will therefore be treated for their alcohol problems after fewer years of excessive consumption than men (Schuckit et al., 1995).

Concerning psychological and psychiatric aspects, psychiatric co-morbidity is

more common among alcohol-dependent women than men (Helzer & Pryzbeck,

1988; Kessler et al., 1997). In the National Comorbidity Study (NCS), Kessler et

al. (1997) found that alcohol-dependent women had compared to men more

lifetime anxiety (61 % versus 36 %) and more lifetime affective disorders (54 %

versus 28 %) whereas alcohol-dependent men had more antisocial personality

disorder (8 % versus 17 %). Furthermore, it is more common that the onset of

psychopathology precedes the abuse in women than in men, with exception of antisocial personality disorder (Hesselbrock, Meyer & Keener, 1985; Kessler et al., 1997). Concerning risk factors in developing alcohol abuse or dependence, it was found in a Swedish study (Spak, Spak & Allebeck, 1997), that women with earlier psychiatric and/or psychological problems, early deviant behaviour, alcohol intoxication before the age of fifteen and a history of sexual abuse before the age of thirteen, had elevated risks of developing alcohol abuse or dependence.

Drinking patterns

There have been several ways to categorise individuals with alcohol problems according to their drinking patterns. Jellinek’s typology (1960) of five types of alcoholics (1) alpha, (2) beta, (3) gamma, (4) delta and (5) epsilon are an example of a typology where drinking patterns are the determining factor that distinguish different sub-groups. His hypothesis was that alpha and beta types did not have a physical dependence and therefore, these types of alcoholics were not having an alcoholism disease. Alpha types were drinking due to the relief of bodily or emotional pain while beta drinkers had a controlled but high

consumption, which may lead to physical diseases such as gastritis and cirrhosis.

The true alcoholics were gamma and delta alcoholics, characterized by increased tolerance to alcohol, adaptive cell metabolism, withdrawal symptoms and

“craving”. The differences between gamma and delta types were that gamma

types had loss of control whereas delta types had an inability to abstain. Epsilon

drinkers or the periodic drinkers were the least known of the alcoholic types and

Jellinek (1960) was not sure if this specific type had a disease according to the

above stated criteria.

A more recent typology that also is based on different drinking patterns is the so called Lesch typology (Lesch & Walter, 1996). This was derived from a long- term prospective study (18 years) and differentiates four types of alcoholics:

Type 1 has severe withdrawal and mainly a somatic dependence. In these individuals, social drinking develops to habitual drinking and abstinence

symptoms occur early. Often they have a positive family history for alcoholism.

The alcohol consumption of type II is mostly a coping strategy against anxiety.

These individuals are frequently becoming aggressive when they are intoxicated but they have no severe somatic alcohol-related disorder or withdrawal. Type III is using alcohol as a self-medication against depression or sleep disorder. These individuals have often psychiatric co-morbidity, aggressive behaviour even without alcohol and mild somatic withdrawal. Type IV has a cerebral

disturbance or prenatal damage before the termination of brain development.

These individuals often have grand mal seizures (not only during withdrawal) and deterioration of there psychic, organic and social sphere (Lesch & Walter, 1996; Pombo & Lesch, 2008).

Some single drinking pattern variables have been of specific interest since they have been able to e.g. predict future alcohol problems. One of the most

important single variables, namely the age of which drinking first time occur, have constantly been regarded as an important variable for predicting the future pattern of alcohol use as well as other psychosocial problems. Early onset of drinking in late childhood/early adolescence is thus associated with increased risks of failure in school, childhood psychiatric disorders, criminality, deviant behaviour, and an overall lowered life-satisfaction (McGue, Iacono, Legrande, Malone & Elkins, 2001). Additionally, these early onset drinkers have increased risks of developing alcohol-related problems in adulthood (Chou & Pickering, 1992; DeWit, Adlaf, Offord & Ogborne, 2000; Grant & Dawson, 1997).

However, despite increased risks of consuming alcohol in early ages (defined as

the age of 14 years or earlier) the majority of these individuals do not develop problems later on (Grant & Dawson, 1997). Other specific risk-factors which are associated with later drinking problems include being male, feeling drunk at initiation of drinking, having parents with alcohol-dependence and having criminal behaviour (Warner & White, 2003). The antecedents for early onset of drinking seems more to be environmental than genetic, where for example peer influences, availability of alcohol and sibling interactions are important factors (Kuperman et al., 2005; Rhee, Hewitt, Young, Corley & Stailings, 2003; Rose, Dick, Viken, Pulkkinen & Kaprio, 2001).

Psychiatric co-morbidity

The co-morbidity of alcoholism and psychiatric disorders encompass a diverse spectrum of problems and diagnoses and therefore individuals with these

problems have different needs and require different types of treatment (Baigent, 2005). Psychiatric disorders are on the average twice as common among

individuals with alcohol problems than the rest of the population (Helzer and Pryzbeck, 1988). Alcohol-dependent individuals, especially with severe mental illness as for example schizophrenia, have on the average more severe problems in other life-areas than individuals without co-morbidity (Brunette, Mueser &

Drake, 2004). Individuals with schizophrenia or other severe mental illnesses and an additional substance misuse are associated with a wide range of negative outcomes, such as increased rates of homelessness, legal problems, violence, treatment non-compliance, HIV-infection and family stress (Drake & Brunette, 1998).

It should be noted that treating individuals for their alcoholism, without taken in

consideration a possible psychiatric disorder, may lead to adverse treatment

outcome (Öjehagen, Berglund, Appel & Skjaerris, 1991). Despite the fact that there is higher prevalence of psychiatric disorders among alcohol-dependent individuals than in the rest of the population, the etiologies for this higher

prevalence are still unknown (Mueser, Drake & Wallach, 1998). One suggestion is that it is more common that the psychiatric disorder precedes the alcohol misuse (Berglund & Öjehagen, 1998). In the following, there will be a short presentation of some of the most common psychiatric disorders which are related to alcoholism: antisocial personality disorder, depression and anxiety (Berglund & Öjehagen, 1998)

Antisocial Personality Disorder

The relation between antisocial personality disorder (APD) and alcohol- dependence is well documented (Waldman & Slutske, 2000). In the

Epidemiologic Catchment Area Study (ECA), the risk of having APD was 21 times higher in individuals with alcohol-dependence than for the rest of the population (Helzer & Pryzbeck, 1988). Individuals with APD and alcohol- dependence have consistently been found to have a familial history of

alcoholism, earlier age at onset of alcohol-dependence, more severe physical dependence, abuse of other drugs, more adverse physical, social and legal consequences and worse prognosis for recovery (Babor, 1996; Cadoret, Troughton & Widmer, 1984; Hesselbrock et al., 1985). The relation between criminality, substance abuse and APD is strong, also in longitudinal studies (Fridell, Hesse & Billsten, 2008; Fridell, Hesse, Meier-Jaeger & Kühlhorn, 2008).

The causes of co-morbidity between APD and alcohol-dependence are still

unknown, but there are three basic hypotheses: 1) they co-occur, because they

share common causes, 2) APD causes alcohol-dependence or 3) alcohol-

dependence causes APD (Waldman & Slutske, 2000). The results from different studies reveal that simple unidirectional models may not adequately explain the association between APD and alcohol-dependence. Therefore, models of

reciprocal causation between APD and alcohol-dependence may need to be considered (Waldman & Slutske, 2000).

Depression

Clinical and epidemiologic studies as well as twin, adoptive and family studies have revealed that there is a relation between alcohol-dependence and

depression. In clinical samples, Merikangas and Gelernter (1990) found that the frequency of depressive symptomatology range from 16 to 69 percent in

hospitalized alcoholics. In the National Comorbidity Study (NCS) the risk of lifetime occurrence of major depression in alcohol-dependent individuals was two-folded compared to the general population (Kessler et al., 1997).

Concerning the etiology of alcohol- dependence and depression, a common suggestion is that these are not manifestations of the same underlying disorder and that they do not share a common etiology (Swendsen & Merikangas, 2000).

Although the relationship between depression and alcohol-dependence is complex and should be interpreted with caution, there is more support for the conclusion that alcohol-dependence more often generates depression than the contrary. This may be explained by the fact that alcohol-dependence cause severe impairment in social, health and occupational domains and these stressors, which can be both acute and chronic, have been consistently

implicated in the etiology of depression (Brown & Harris, 1989). It may also

e.g. be due to the impairment of central serotonergic neurotransmission

following long-term alcohol consumption (Berggren, Eriksson, Fahlke &

Balldin, 2002). Also important, especially in clinical settings, is that some psychiatric symptoms are temporary substance induced symptoms, that have occurred during intoxication or withdrawal conditions (Balldin et al., 1992a;

Fahlke, Berggren, Lundborg & Balldin, 1999).

Anxiety Disorders

Prevalence data from the ECA-study (Regier et al., 1990) and the NCS-study (Kessler al., 1997) reveal that the risk of having an anxiety disorder is at least doubled in alcohol-dependent individuals, with an exception for simple phobia.

Thus, these data suggest that there is a significant relationship between alcohol- dependence and nearly all forms of anxiety disorders.

Regarding the etiology of co-morbidity between alcohol-dependence and anxiety disorder, either anxiety disorder or alcohol-dependence can serve as causal stimulus for the other (Kushner, Abrams & Borchardt, 2000). However, the best way of understanding the etiology of co-morbidity between these two disorders, relying on pharmacological and behavioural laboratory findings, is to consider that there is an interaction between anxiolytic and anxiogenic processes in alcohol-dependence. This is explained by a feed-forward cycle that emerges:

Drinking is reinforced by short-term anxiety reduction, but is followed by

anxiety induction on withdrawal (Stockwell, Hodgson & Rankin, 1982).

Personality

Aspects of personality have over the past 50 years had a great influence on alcohol research when explaining the etiology of alcohol-dependence. In the first edition of Diagnostic and Statistical Manual of Mental Disorders (DSM-I;

American Psychiatric Association, 1952) alcoholism was even considered as a personality disorder. Today, the general conclusion among researchers is that there is no specific personality profile that predicts future alcoholism. Instead, the common assumption is that certain personality traits or characters may enhance the risk of developing problems with excessive alcohol intake. Not to forget is also the possibility that alcoholism per se may change the personality.

Thus, personality traits which enhance the risk of future alcoholism are not necessarily synonymous with personality traits that are most commonly seen in clinical alcoholics (Barnes, 1983). Regarding etiological aspects of personality and alcoholism it is argued that personality variables should be seen in the nexus of other causal variables in more complex models (Sher & Trull, 1994) where personality variables can act as either mediators or moderators. Figure 1 illustrates an example of how the personality can act as a mediator. Thus, the personality is one possible variable that can explain (mediate) why some individuals with a family history of alcoholism develop alcoholism later in life (e.g. Sher, 1991).

Figure 1. An example of a mediator

positive Personality

(i.e. mediator) Alcoholism

Personality can also act as a moderator, as seen in figure 2. This means that personality can have a moderating effect when it interacts with other risk-factors and exacerbates the likelihood of developing excessive drinking or alcoholism.

The following is an example of this: Expectancies are moderated by personality in the outcome on adverse drinking. Thus, risk-taking adolescents may have a more adverse drinking outcome (e.g. binge drinking) than other adolescents because their expectancies are interacting with their risk-taking personality which do not consider the risks of adverse drinking (Del Boca, Darkes, Goldman

& Smith, 2002)

Figure 2. An example of a moderator

There are several trait theories of personality. One accepted theory, which has also been implicated in alcohol research, is the Eysencks “Big three” model (Eysenck, 1990). The underlying dimensions of Eysencks model are

neuroticism, extraversion and psychoticism. Another theory that has been widely used in alcohol research is Cloninger’s (Cloninger, Svrakic & Pryzbeck, 1993) psychobiological theory of temperament and character personality

dimensions. This theory is partly based on a neurobiological perspective of personality and has been frequently used for studying the relationship between

Expectancies

Personality (i.e. moderator)

Adverse drinking

neurobiology and personality. The temperament dimensions in Cloninger’s model are novelty seeking, harm avoidance, reward dependence and persistence (Cloninger et al., 1993). Two of these personality dimensions (novelty seeking and harm avoidance) are in some aspects similar to Eysencks personality dimensions psychoticism and neuroticism. For example high levels of novelty seeking as well as psychoticism are characterised by impulsiveness whereas low levels of these dimensions are characterised by a calm and thoughtful temper.

High levels of neuroticism and harm avoidance are characterised by anxiousness whereas low levels are characterised by emotional stability. Since the

dimensions of neuroticism/harm avoidance and psychoticism/novelty seeking are the dimensions that are most related to alcoholism, they are further presented below.

Neuroticism/Harm Avoidance

A number of studies have confirmed that the dimension of neuroticism is related to clinical alcoholism (e.g. Kessler et al., 1997; Kushner et al., 1996). What is less clear is if this dimension also is associated with the risk for future alcohol problems. Several longitudinal studies do not suggest that neuroticism has a strong causal role for alcoholism (e.g. Vaillant & Milofsky, 1982). However, other longitudinal studies do suggest such a causal role, but only in late onset alcoholics (e.g. Cloninger, Sigvardsson & Bohman, 1988). Our understanding of the etiological aspects of this dimension is thus far from complete. One

explanation of this inconsistency is that there are mediators and moderators that

can influence the relationship. For example, motives and expectancies are

factors that can act as either mediators or moderators between personality and

alcohol use. Regarding motives, one motive to drink is to regulate an emotional

state. Motives as drinking to “cope” have been associated with neuroticism

(Cooper, Frone, Russell & Mudar, 1995). Expectancies can be explained as anticipators of behaviour. One hypothesis of the relationship between

neuroticism and alcohol use is that expectancies moderate behaviour (Kushner, Sher, Wood & Wood, 1994). Thus, those neurotic individuals who believe that alcohol can reduce tension and anxious feelings are more likely to drink.

The relationship between neuroticism and alcoholism is thus complicated since there are other possible variables, for example motives and expectancies that may influence this relationship. More prospective studies, as well as more complex analysis, are therefore needed for better understanding of these relations.

Psychoticism/Novelty Seeking

There is strong evidence that aspects of this dimension, for example impulsiveness and disinhibition, are associated with future alcoholism.

Alcoholics score higher in this personality dimension than the normal population (e.g. Bergman & Brismar, 1994). Also prospective studies consequently point out that traits relating to this personality dimension (e.g. deviant behaviour, aggression, tough-mindedness) are associated with future alcohol problems (e.g.

Caspi, Moffitt, Newman & Silva, 1996; Cloninger et al., 1988). High scores in this personality dimension have also strong associations with the APD

(discussed more in detail in previous part of this introduction) and some propose that instead of indexing a personality dimension it represents a developmental disorder that manifests itself in the form of alcoholism (Sher, Trull, Bartholow

& Vieth, 1999).

The sensation-seeking personality construct (Zuckerman, 1979) involves four components: thrill and adventure seeking, experience seeking,

impulsiveness/disinhibition and boredom susceptibility. This personality construct have in numerous studies been associated with excessive alcohol intake (e.g. Alterman et al., 1990). However, in a meta-analysis where 61 studies were analysed the component that was most associated with excessive alcohol intake was the impulsiveness/disinhibition component (Hittner & Swickert, 2006). Thus, that study concludes that impulsiveness/disinhibition is the trait that best explains the association between sensation-seeking and excessive alcohol intake.

Personality, alcoholism and serotonin

One monoamine that has been related to aspects of personality, behaviour and alcoholism is serotonin (5-hydroxy-tryptamine). Serotonin, which is widely distributed neurotransmitter in the brain, is represented in several different brain areas including the anterior cingulate cortex, amygdala, hippocampus, ventral striatum, hypothalamus and many other interconnected structures (Lesch, 2007).

Serotonin is among others involved in the regulation of emotions and also behaviour such as food consumption and digestion, relaxation, growth, sleep, passivity and inactivity (Feldman, Meyer & Quenzer, 1997). Baumgarten and Grozdanovic (1995) suggest that the serotonergic neurotransmission is involved in a “protective filter effect” reducing the impact of sensory information. Thus, according to the above mentioned hypotheses, central serotonergic

neurotransmission is involved in promoting feelings of calmness and security.

Several studies have found that reduced central serotonergic neurotransmission

is correlated to impulsive and aggressive behaviour but also to negative mood

states such as anxiety and depression. In non-human primate studies, it has been found that low serotonergic neurotransmission predisposed subjects to both aggressive behaviour and excessive alcohol intake (Heinz et al., 1998). In humans, the so called “type 2” alcoholics have been found to have low

serotonergic neurotransmission (Fils-Aime et. al., 1996). Type 2 alcoholics, who are described in detail in the forthcoming, are among others characterised by early onset alcohol consumption, aggressive and anti-social behaviour

(Cloninger, Sigvardsson & Bohman, 1996). Since low serotonergic

nerurotransmission seems to be related to both aggressive behaviour as well as anxiousness and depressed mood (e.g. Knutson et al., 1998), one hypothesis is that low serotonergic neurotransmission and aggressive behaviour is mediated by feelings of insecurity and threat. Virkunnen et al. (1994) have found that alcoholics with high aggressiveness and low serotonergic neurotransmission also had high anxiety levels. Thus, threats and insecurity are variables that in this hypothesis explain why some alcoholics with low serotonin neurotransmission, and with symptoms of anxiety and depression, also behave aggressively.

Another aspect of the relation between reduced central neurotransmission and alcoholism is that long-term excessive alcohol drinking per se may reduce central serotonergic neurotransmission. For example, in a study by Berggren et al. (2002) a significant relationship between years of excessive alcohol

consumption and serotonergic neurotransmission was found; thus, the longer duration of excessive alcohol intake, the lower serotonergic neurotransmission.

One possible explanation for this relationship is that alcohol has neurotoxic

effects on serotonergic neurons. Thus, low serotonergic neurotransmission in

alcoholics may be explained by a neurotoxic effect in serotonergic neurons

caused by long-term heavy alcohol consumption. On the other hand, it is also

possible that there is a pre-existing reduction in serotonergic neurotransmission.

Heredity

Alcoholism is more common in some families than others and it is well

established that there are higher risks of developing alcohol-related problems or alcohol-dependence in families with a history of alcoholism (Cotton, 1979). The heritability of alcohol consumption is estimated to be 35-40 percent in twin studies (Clifford & Hopper, 1984; Kaprio et al., 1987; Partanen, Bruun &

Markkanen, 1966). Children of alcoholics have, compared to other children, higher risk for a variety of problems that may include behavioural,

psychological, cognitive or neuropsychological deficits (Johnson & Leff, 1999) and also more risk-related physiologic responses to alcohol, i.e. a lack of

alcohol’s intoxicating effects (Schuckit & Smith, 2000). However, such characteristics are not unique to children of alcoholics and these factors can mediate risk of developing alcohol problems in children without a family history too (Molina, Chassin & Curran, 1994). Despite the higher risk for individuals to develop alcoholism if they have a family history of alcoholism, the majority of these individuals do not develop alcoholism. One explanation for this is that the expression of familial alcoholism is very different. When comparing children of families without alcoholism, families with alcoholism and families with

alcoholism and antisocial personality disorder, it has been shown that children

from families with both alcoholism and antisocial disorder have the highest risk

levels for developing alcohol problems (Zucker, Ellis, Bingham & Fitzgerald,

1996). In a recent study regarding personality traits as familial risk factors for

developing alcohol-dependence, it was found that high levels of novelty seeking,

characterised as impulsive behaviour, increased the risk of developing alcohol-

dependence in high risk families (Grucza et al., 2006). High risk families were

in this study characterised as families with either one or both of the parents

having alcohol-dependence. Furthermore, low degree of novelty seeking

behaviour in individuals from high risk families acted as a protecting personality trait against development of alcohol-dependence (Grucza et al., 2006).

Type 1 and type 2 alcoholism

The typology of type 1 and type 2 alcoholism is based on different heredity patterns. It was developed in the beginning of the 1980´s by the American psychiatrist Robert C. Cloninger and his Swedish colleagues. They studied a large number of Swedish adoptees and their biological and adoptive parents (Cloninger, Bohman & Sigvardsson, 1981). The adoptees comprised all sons, born between 1930 and 1949 to single women in the city of Stockholm, and adopted by non-relatives at early ages. Their biological fathers were known. The adoptees were studied from early ages and up to adult ages (23-43 years). The researchers used data from the social system, the criminal registry, local

agencies of the national health insurance, hospital records and so on. The aim of the study was to find out whether there is a way to distinguish alcoholics

according to their inheritance patterns of alcoholism and to study the relative contributions of genetic and environmental factors in the individual’s

susceptibility to develop alcoholism. The adoptees which had developed alcoholism fell into two groups according to their drinking patterns and the characteristics of their biological parents; especially the biological father’s drinking behaviour (Cloninger et al., 1981).

One group, labelled type 1 alcoholics, included males with mild or in some cases severe alcoholism. This type of alcoholism developed during adulthood. It was characterised by binge drinking that relatively rapidly progressed from mild to severe alcohol abuse. They experienced loss of control when drinking

followed by excessive feelings of guilt. The genetic predisposition (i.e.

alcoholism in the biological parents) contributed only slightly to this category. If the biological parent of the adoptee yet had alcoholism, this was a mild form of alcoholism which did not require treatment, had started in adult years and was not associated with criminality. There was also an environmental influence in the type 1 individuals, indicating that if the non-biological father of the adoptee had alcohol problems there was also an enhanced risk of developing alcoholism in the adoptee.

The second group, type 2 alcoholics, were characterised by moderate alcohol abuse compared to type 1. This type of alcoholism developed during

adolescence or early adulthood. Alcohol consumption was frequently accompanied by fighting and arrests. The genetic predisposition was much stronger for these adoptees with biological fathers having severe alcoholism which had started during adolescence and that required treatment. The biological fathers were also frequently involved in criminality. Bad environmental

influences aggravated the severity of alcoholism but did not affect the

frequency. In the study (Cloninger et al., 1981), it was also found that the type 2 alcoholism was much less prevalent compared to type 1 alcoholism. But if an adoptee was predisposed to type 2 alcoholism there were significantly higher risk of becoming an alcoholic. These findings have later been replicated in a similar study where also women were included (Sigvardsson, Bohman &

Cloninger, 1996).

The conclusion according to these two studies (Cloninger et al., 1981;

Sigvardsson et al., 1996), is that type 1 alcoholism can be both mild and severe, affects both men and women and starts after years of heavy drinking in adult life (> 25 years of age). The predisposition can be either genetic or environmental.

The type 2 alcoholism affects only sons of male alcoholics, has a strong genetic

predisposition and is influenced only weakly by environmental factors, develops

in adolescent years or early adulthood (< 25 years of age) and is often associated with criminality (Cloninger et al., 1996).

Statistically developed categorisations

In the beginning of the 1970´s, some researchers started to incorporate more complexity into their categorisation models. This was made either by postulating subtypes that encompass multiple defining characteristics or by deriving

typological characteristics from empirical data. This more sophisticated way in typology research was carried out by new statistical methods, such as

multivariate analysis.

Type A and type B alcoholism

The perhaps most established statistically developed categorisation of alcohol- dependent individuals is the typology of type A and type B alcoholism (Babor et al., 1992). In the beginning of 1990, the American psychiatrist Thomas Babor and his colleagues tried to find a typology with good construct, discriminative and predictive validity (Babor et al., 1992). Construct validity means that a theoretical construct fits statistical relationships that are observed empirically. If subgroups can be clearly distinguished from each other in terms of different characteristics, then the typology has discriminative validity and if typology can predict treatment outcomes for different subgroups, then the typology has

predictive validity.

Based on previous typology research, Babor and his research-group included 17

theoretically relevant indicators of premorbid risk factors, pathological use of alcohol and other substances, severity and consequences of drinking and psychiatric symptoms. The subjects in the study were 321 alcohol-dependent patient volunteers, recruited from different treatment centres in the United States. According to the above stated criterion, Babor and colleagues (1992) decided that a two-cluster solution was the best way to categorise the subjects.

Consequently, this solution produced two clusters of alcoholics that were

significantly differentiated in all 17 variables in male alcoholics and in 13 of the 17 variables in female alcoholics, indicating that the typology had good

construct validity. The groups were then compared according to other

characteristic variables such as personality variables, consequences of drinking and childhood aggressive behaviour. Several differences were found between the two groups, indicating that the typology also had good discriminative validity. The typology also had good predictive validity shown by different patterns of treatment outcome in different groups. Specifically, type A had a more favourable response to interactive therapy whereas type B responded better to coping skills therapy. Babor and his colleagues (1992) labelled the two

groups type A and type B, whose alphabetical designation derived from the Greek and Roman mythology of Apollo and Bacchus. The god Apollo was, according to the Greek mythology, intellectual, artistic, creative and self- restrained. On the contrary, the god Dionysos (Bacchus in Roman mythology) was reknown for drunken revelry, sexual abandonment and physical aggression.

According to the results of this study (Babor et al., 1992), type A was

characterised by later onset of alcohol problems, fewer childhood risk factors,

less severe dependence, fewer alcohol-related physical and social consequences,

less psychopathological dysfunction and less distress in the areas of work and

family. The type B group was characterised by childhood and familial risk

factors, earlier onset, greater severity of dependence, other substance abuse and

more serious consequences of drinking and more co-occurring psychiatric disorders.

Other categorisations

Several new statistically developed categorisations have been added to the international literature in recent years. One example of this is Windle and Scheidt’s (2004) typology. They have identified four sub-types of alcoholism.

Mild course: Individuals in this category have later alcohol onset, lower levels of alcohol consumption, withdrawal symptoms, few childhood conduct

problems and low level of family history of alcoholism. Polydrug: In this

category individuals have the highest levels of polydrug use and benzodiazepine

use. Negative affect: In this category, many individuals have symptoms for

major depressive disorder and generalized anxiety disorder. They have also the

highest number of psychopathy symptoms. Chronic/antisocial personality

symptoms: Individuals in this category have the highest levels of alcohol

consumption and impairment (adverse social consequences and dependence

severity). They have also the highest levels of adult antisocial behaviour. For the

categorization of the individuals, Windle and Scheidt (2004) used a cluster

analytical technique in 802 alcohol-dependent individuals admitted to an

inpatient treatment.

Socially stable alcoholics - what characterises them?

It is obvious that alcoholism has several features and that the complexity relies among others on the multifactorial etiology and the different characteristics that these individuals are comprehended by, e.g. drinking patterns, psychiatric co- morbidity, personality and heredity. It should be noted however that these characteristics as well as categorisations mainly are based on data from clinical samples and more seldom on non-clinical samples (Del Boca, 2004).

Concerning non-clinical samples, their primary problem is their excessive alcohol intake, not severe psychiatric co-morbidity, unemployment, criminal record or even homelessness (Raimo, Daeppen, Smith, Danko & Schuckit, 1999). Moreover, it is more common that they are socially stable. Being socially stable or well adjusted can be both a constraining as well as a facilitating factor in different stages of alcoholism. Regarding the process of help-seeking, social stability was in one study constraining factors for seeking help (Hajema, Knibbe

& Drop, 1999) irrespective of the severity of problem-drinking. However, in studies of treatment outcome, social stability facilitates a successful treatment outcome (Wallace, McNeill, Gilfillan, MacLean & Fanella, 1988; Öjehagen, Skjaerris & Berglund, 1988). Moreover, poor socio-demographic factors are predictors of relapse in detoxified alcoholics (Walter et al., 2006).

The definitions of social stability have varied somewhat in different studies. For

example, in two studies investigating treatment outcome, social stability was

defined as being cohabitant and employed (Öjehagen et al., 1988). In the study

of Hajema et al. (1999), social stability was defined as employment and somatic

health. Chanraud et al. (2007) investigated brain morphometry and cognitive

performance in a sample entitled alcohol-dependent individuals with preserved

psychosocial functioning. The motive for this definition was that the individuals

scored high in a scale measuring social adjustment which in this scale was

represented by degree of social and leisure activities, relationships with partner and children and economic status (Chanraud et al., 2007). In a study by

Nordström and Berglund (1987), investigating long-term adjustment in alcohol- dependence, good social adjustment was defined as at most 30 sickness cash benefit days every year and a sickness cash benefit extending 50 Swedish

crones/day (this excluded part-time employers). In the study by Schuckit, Smith and Landi (2000), investigating the 5-year clinical course of “high functioning”

men with alcohol use disorders, high functioning was regarded as education, employment and cohabitation.

According to the Swedish National Institute of Public Health (2008a), about 17 percentages of men and 10 percentages of women have alcohol problems. It can be assumed that the majority of these individuals are socially stable, e.g. having employment and a permanent residence, since these figures derives from a survey study where a sample of the Swedish population between 16-84 years were randomly selected. Moreover it is also likely that some of these individuals have never been in contact with treatment centres for their problems. Therefore we need more knowledge about what characterise individuals with alcohol problems that still are socially stable. Thus, the main theme in this thesis was to investigate socially stable alcoholics. The superior definition of social stability in this thesis is that the individuals are a part of the labour market and that they have a permanent residence. This definition is broad and it comprises a

heterogeneous group of individuals. Nevertheless, the intention with this definition is that it excludes individuals with problematic psychosocial

adjustment, as for example individuals with severe mental illness or individuals with severe criminal behaviour, i.e. groups of alcohol-dependent individuals that have other requirements.

Aims

People who abuse alcohol are a heterogeneous group. One major effort in the field of alcohol research has been to study different groups of excessive drinkers in order to study possible pathways into alcohol use disorders and the long-term course of alcohol-dependence. The general aim of this thesis was to investigate socially stable alcoholics regarding personality, physical and mental health and drinking patterns. Variables that were investigated in relation to the above mentioned aspects were demography, age, gender, earlier treatment experience and central serotonergic neurotransmission. Socially stable alcoholics are

important to study since this group may constitute the majority of all individuals with alcohol problems.

This thesis includes four studies and the specific aim of each study was the following:

Study I aimed to describe demographic and clinical characteristics in male individuals recruited by advertisements and to compare these individuals in relation to their prior experience of treatment for alcoholism.

Study II aimed to investigate personality dimensions in male individuals, in relation to central serotonergic neurotransmission and history of excessive alcohol consumption.

Study III aimed to further explore the personality dimensions of male alcoholics and also compare them to a population-based male control group.

Study IV aimed to investigate health functioning and drinking patters in a

Swedish treatment sample in relation to age and gender.

Methods

Participants

In study I, II and III, individuals were recruited by advertisements in a Swedish regional daily newspaper between the years 1988 -1996. The main purpose of the advertisements was to recruit individuals for investigating different

pharmacotherapeutical interventions on excessive alcohol intake (see Balldin et al., 1994; Balldin et al., 2003; Eriksson, Berggren, Blennow, Fahlke & Balldin, 2001a; Eriksson, et al., 2001b). Inclusion criteria for participation in the studies were that individuals should be males, they should have a permanent residence and they should be between 18-65 years old. Specifically in study II and III the subjects should consume a minimum of 300 grams pure alcohol/week during the last year and they had to meet the DSM-IV (American Psychiatric Association, 1994) criteria for alcohol abuse or alcohol dependence. Furthermore they should not have received diagnoses of abuse or dependence other than alcohol and nicotine. They should also be employed or living on a pension and be without physical or psychiatric disorders not associated with excessive alcohol

consumption.

In study IV, women and men admitted to a Swedish 12-step inpatient treatment,

between 2004 to 2005, were asked to participate in a longitudinal study (the

Göteborg Alcohol Research Project; GARP) evaluating possible psychological

and neurobiological predictors of alcohol-dependence as well as outcome of the

treatment. Inclusion criteria for these individuals was that they had to meet the

DSM-IV criteria for alcohol-dependence (American Psychiatric Association,

1994).

Controls

Out of the four studies in this thesis, study III included a male control group (n = 131). The controls were a sub-group of twin-pairs from the ongoing Swedish Adoption/Twin Study of Ageing (SATSA; Pedersen et al., 1991). This sub- group participated 1995 in a sub-study on the relationship between personality and health (Gustavsson, Weinryb, Göransson, Pedersen & Åsberg, 1997).

Among others they answered a postal survey including a personality questionnaire which was analysed and compared in study III.

Procedures and instruments

Study I

After reading the advertisements in the newspaper, the individuals had to call the investigators for further information of the pharmacotherapeutical

interventions, and if they were interested to participate, a structured telephone interview was performed by psychologists or registered nurses according to a standard protocol. The interview focused on demographic variables, current psychiatric and somatic symptoms, use of medication for psychiatric/somatic symptoms, drinking patterns and previous experience of treatment for excessive alcohol drinking. The data in study I is thus based on questions from the

standardised telephone interviews. A sample of 342 male individuals was

included in the study.

Study II

In this study a sub-group of the individuals from study I (n = 33) was further investigated in the following variables: central serotonergic neurotransmission, personality profile and history of excessive alcohol consumption.

Neuroendocrine challenge test: The way of investigating central serotonergic neurotransmission in these individuals was done by a neuroendocrine challenge test. Blood samples for determination of prolactin were collected at the time- point 0 min and every 60 min thereafter for 4 hours. D-fenfluramine

hydrochloride (Isomeride®) was administered orally in a dose of 30 mg after collection of the blood sample at time 0 min. The hormonal response to serotonin active drugs, such as fenfluramine and selective serotonin reuptake inhibitors, has been proposed to be useful functional tests for central

serotonergic activity in humans. Prolactin is secreted in increasing amount from the pituitary gland when central serotonergic neurons are activated by

fenfluramine, which releases presynaptically located stores of serotonin. Thus, the magnitude of this prolactin-response, which is measured in repeated blood samples after oral administration of fenfluramine, can be used as an indirect measurement of central serotonergic function. For a more detailed discussion of this topic see Eriksson et al. (2006a).

Personality profile and psychiatric assessment: For investigating personality profile, the Temperament and Character Inventory (TCI; Cloninger et al., 1993) and the Karolinska Scales of Personality (KSP; Ramklint & Ekselius, 2003;

Schalling, Åsberg, Edman & Oreland, 1987) were used. TCI assesses seven

personality dimensions, of which four of them are basic dimensions of

temperament: novelty-seeking, harm avoidance, reward dependence and

persistence. The remaining three are character dimensions: self-directedness,